“Let our rigorous testing and reviews be your guidelines to A/V equipment – not marketing slogans”
Facebook Youtube Twitter instagram pinterest

HomePNA, MoCA, HomePlug, UPA Systems

By

The first option is one I have familiarity with, having used it back as far as 2003 when it was the Home Phoneline Networking Alliance. The standard, in version 2.0 at the time with a maximum throughput of 10 Mbits/s, was fast enough for sharing an Internet connection and some PC gaming, but not fast enough for HD streaming duties, not that there was much of that to do back then. Adapters were available both as usb dongles sold by DSL providers and as third party internal PCI computer cards of which I typically adopted the later.

HomePNA is a non-profit industry association with the goal of standardizing home networking using existing wiring.  The current standard, HPNA 3.1, has a maximum throughput of 320 Mbits/s and a range of 300 meters over phone wiring or 1000+ meters over coaxial wiring.  With a QoS mechanism written into the specification to prioritize network transmission and reduce data collisions, HPNA typically achieves about 90% of maximum throughput.  With it roots in phone wiring, HPNA has become something of the standard of choice for telecom based triple play IPTV systems such as AT&T U-verse.  The current network running through my house is mixed, using an HPNA coax connection into the gateway and to the IPTV receivers but is also distributed at the gateway to an Ethernet backbone between floors for networking other devices and computers with several supplemental HomePlug Turbo connections (as described below).

Up through the HPNA 3.0 specification, the standard covered networking over phone lines but the most recent revision to HPNA 3.1 includes coaxial cable networking.  HomePNA operates at frequencies above voice, DSL, and ISDN on phone and coaxial wire, but below those used for broadcast TV and direct broadcast satellite (DBS) TV on coaxial wiring to avoid interference.  Cable based IP standard DOCSIS overlaps with HPNA frequencies, making the two incompatible for home networking on coaxial.  Supported by a wide array of companies, HPNA 3.1 is also the current home networking standard for phone line and coaxial cable networking recommended by the International Telecommunication Union, Telecommunication Standardization Sector (ITU-T) under Recommendation G.9954 (01/07): Home networking transceivers - Enhanced physical, media access, and link layer specifications.

To use HPNA, a gateway and/or bridges will be required with gateways typically supplied as part of an ISP service.  HPNA also can be set up without an ISP gateway simply by connecting at least two bridges, one at the central network node or primary Internet service connection, and one to any outlying node.  From the bridges, any Ethernet enabled device can be connected directly to the bridge or the bridge can be connected to a switch to allow multiple connections.  Network setup is as simple as plugging the bridges into the appropriate wire, phone or coax, and plugging in an Ethernet cable with no software or drivers to set up on any of the connected machines.

HPNA products such as bridges, IPTV set top boxes, and gateways are available from a number of member companies listed on the HPNA website.  While most of the gear would come with a particular service provider, bridges can be used on any compatible home network and look to be around $80 or less at street prices. 

MoCA

The Multimedia over Coax Alliance is another standards group that looks to make use of existing wiring commonly found in most homes.  The latest standard is MoCA 1.1 that pushes 175 Mbits/s over existing home coaxial wiring at frequencies around 1 GHz.

Also supported by an array of companies, some overlapping with the HPNA group, MoCA has been incorporated in recommendations by the Digital Living Network Alliance, a cross industry group that includes consumer electronics, computing, and electronics to promote cross communication.  As evidenced by the member company list, MoCA tends to be the preferred standard of cable based TV triple play providers.

Generally, good MoCA performance is easily achievable, but certain circumstances will affect data rates.  Engadget HD found issue with video signal splitters, common in many TV installations, inline between MoCA bridges being a serious detriment.  Four port splitters are common where service enters a home, but as a worst case, they encountered a 16-port splitter that reduced throughput to 30 Mbits/s.  Odds are that HPNA over coaxial will suffer similar difficulty with inline video splitters.  Poor performance may require a bit of clean up to eliminate any unnecessary or junk hardware on the line.

Using MoCA is similar to HPNA with service providers supplying gateways that can be used as the main network interface at a central node, and one or more bridges required to connect Ethernet nodes to the coaxial backbone.  Bridges run around $200 a pair, but are easy to use and provide enough bandwidth with headroom to stream HD media and still have room for other network and Internet connections.  The coaxial bridges will pass through ATSC broadcast and cable TV signals but not satellite TV.  The MoCA system is also not compatible with HPNA simultaneously on coaxial wiring, but will obviously not have any conflicts with HPNA over phone lines.

Engadget HD has a review of a Netgear bridge and currently available products are listed on the MoCA website.

HomePlug Powerline Alliance

HomePlug is one of several groups looking to make use of existing copper wiring used for power distribution that typically runs to every single room of a house.  This makes it somewhat more convenient than using phone or coaxial wiring, neither of which is as wide spread as power outlets in most homes.  The downside is, that unlike the other mediums that were designed for communication, electrical wiring was not, with the lack of isolation making performance results more dependent on random power line signal to noise ratios (SNR), which varies with wiring quality, signal path, and whatever else is being used on the line.

The original HomePlug 1.0 specification released in 2001 allowed for speeds up to a gross bit rate of 14 Mbit/s in half-duplex mode while the more recent HomePlug AV specification released in 2005 upped speeds to a 200 Mbits/s gross bit rate.  There is also a HomePlug Turbo specification, which at 85 Mbits/s, falls between the other two standards.  The current AV standard claims a minimum QoS for 2 HD video streams, or about 40 Mbits/s over noisy lines.

HomePlug has also been bolstered by adoption as of May 2008 into the Telecommunications Industry Association (TIA) TIA-1113.  This document supports the MoCA 1.0 standard, giving it an American National Standards Institute (ANSI) backed accreditation.

Use of home plug is as simple as plugging in a minimum of two adapters, one connected to the central node or Internet service, and one or more at any outlying nodes.  Again, a single networked device can be plugged into the Ethernet port or a switch can be added for multiple connections.  However, while HomePlug is convenient, there are several issues to be considered.

First is that is that the evolution of the standard does not require backwards compatibility, only that they must be able to coexist on the same power line network.  What this means is that one can have both HomePlug Turbo and AV connections on the same electrical wiring, but each type needs at least two adaptors, one of each connected to the central node or ISP connection.  Outlying nodes can then be of either type as long as the Turbo and AV specified adaptors have an Ethernet bridge at the central node to act as a middleman to allow cross communication.

Second is security, which like Wi-Fi is not automatic and may be an issue in certain circumstances.  The range of HomePlug signals, up to 200 meters, can extend beyond a home’s internal wiring, making it is possible for anyone using HomePlug compliant gear to access a home network or to inadvertently merge with a neighbor’s HomePlug network.

Like Wi-Fi, to simplify initial setup, HomePlug modules default to a standard network password/encryption key and automatically link to all devices in range with the same password/encryption.  Setting up a secure network will require downloading a driver/utility program from the manufacturer to access the adapters and change the password.  HomePlug security is two-tiered using this network password and physical security.  A common network password is set that is hashed into an encryption key shared between all the adapters.  Any HomePlug with a different password/encryption key will then be excluded.  Changing the network password of each module requires either a direct physical connection or the use of a hardwired password unique to each module.  The network password should be set for remote adapters first as they will, one by one, disappear from the network formed with the node directly connected to the computer.  Once all the remote adapters are changed, change the local adapter, restart the utility program and the network will reappear based on the new password/encryption.

The final issue is susceptibility to line noise and signal path that, like interference for Wi-Fi, can rapidly reduce useable bandwidth.  HomePlug field tests performed when the 1.0 standard were current found size and age of a house were significant factors affecting transmission line quality.  Performance results in actual use can vary widely with HomePlug adapters on the same circuit performing better than over several circuits that have to loop back through the fuse box/circuit breaker.  In personal use, I am able to achieve bit rates of 38.9 Mbits/s as the slowest and 82.7 Mbits/s as the fastest connections between two of the three HomePlug Turbo devices currently in my home.  Others have reported worse results lending support to the sensitivity of this type of networking connection to line quality.

Various HomePlug products that are available are listed on the HomePlug website products page and I have compiled a list of product reviews:

Universal Powerline Association

UPA is another of the several competing but similar powerline networking standards.  Overall, UPA performance is similar to HomePlug AV on paper, although HomePlug contends better performance at various SNR over UPA, but a UPA product review I found indicates better UPA performance than with several HomePlug products tested by the same reviewer.

Either standard is probably just as suitable as the other is for home networking purposes and price may be the deciding factor, but they are not directly compatible and at present and may not be able to coexist on the same line.  The possibility of coexistence will be dependant on an upcoming IEEE standard that various powerline networking groups are participating in; more on that below.  One small advantage for UPA, making it a bit simpler than HomePlug, exists in that there is just a single standard on the market at present; there are no legacy UPA compatibility issues, yet.

Setting up UPA network is very similar to HomePlug and requires two or more adapters, one at the central node, and one at each outlying node to be connected.  The security issues with UPA are also similar to those with HomePlug with the possibility of data transmission beyond the home that will require setting network passwords and encryption keys for assured security.

Available UPA products are listed on the organization's website and I have found a review for one of the products for further reading:

HD-PLC Alliance

The HD-PLC Alliance is a third powerline networking organization started and dominated by Panasonic that appears to mostly operate in Japan.  The organization website does not appear to show any available products, but this may become a valid option in the future if the group looks to promote the standard in other territories. 

Summary of Alternate Networking Technology Performance

As I said earlier, the jumble of sources for networking speeds, glossed over by marketing and imprecise jargon, make it difficult to discern what speeds the various technologies are theoretically capable of producing much less what performance can expected in the real world.  To that end, I have endeavored to sort out and summarize throughputs between ideal theoretical performances and actual usable throughput consistent with the terminology defined above.

Alternate Networking Performance

Standard

Release

Gross

Net

Typical

Performance

Range

Compatibility

 

Date

Bitrate

Bitrate

Throughput

Efficiency

 

Issues

 

 

(Mbits/s)

(Mbits/s)

(Mbits/s)

(%)

(Meters)

 

HPNA 3.1 (phone)

2007

160

128

115

89.8

300

 

HPNA 3.1 (coax)

2007

320

320

285

89.1

1000+

DOCSIS (cable internet)

MoCA 1.1 (coax)

2007

n/a

175

100

57.1

n/a

DBS (satellite)

HomePlug AV

2005

200

150

100

66.7

200

Decreases rapidly with SNR

HomePlug Turbo

2004

85

n/a

30

n/a

200

Decreases rapidly with SNR

HomePlug 1.0

2001

14

8

5

62.5

200

Decreases rapidly with SNR

UPA Digital Home 1.0

2006

200

135

100

74.1

200

Decreases rapidly with SNR

As I said previously, typical throughput/goodput will vary depending on particulars of an actual network installation, but phoneline and coaxial cable methods should be the most consistent.

General Issues with Networks on Existing Home Wiring

Surge suppressors and line filtering for Ethernet over existing wiring may cause problems, so route network connections around any filtering of phone line, coaxial, and power outlets to avoid problems.  In particular, with powerline networking as susceptible as it is to line noise and signal path, avoid any sort of power strip or extension cord and plug directly into the wall outlet.  Other electronic devices that do not direct or receive network throughput can be connected as normal through suppressors.

Interoperability between HPNA, MoCA, HomePlug, UPA, etcetera does not exist between any of these standards when they are used on the same wiring system.  HPNA over coaxial and MoCA cannot be on the same line.  The various powerline technologies have taken steps that will at least allow coexistence on the same wiring through the IEEE P1901 Draft Standard for Broadband over Power Line Networks: Medium Access Control and Physical Layer Specifications.  However, the suggestion that a standard that allows multiple coexistent but un-interoperable sub standards is an oxymoron that has generated controversy and will likely confuse consumers.

This is not to say a hybrid network cannot be built, but take care when mixing and matching.  Furthermore, it is possible to build a hybrid with a Wi-Fi network.  The HPNA gateway I use includes wireless functionality and a number of the powerline solutions available have Wi-Fi capability built into the wall adapters.

Some additional help with home networking can be found on Microsoft's website. 

The Future

There currently is a push with many of these disparate home wiring network technologies for convergence and interoperability.

One such push is the G.hn standard being developed by the International Telecommunication Union (ITU) and supported by the HomeGrid Forum.  The standard is an attempt to provide networking interoperability across power lines, phone lines, and coaxial cable at improved data rates of a Gbit/s.  The intent is to produce a single piece of silicon that can use any of the existing home wiring mediums to communicate and to include such chips directly in future electronics with the ultimate goal of leading the standard to smart grid energy management functionality.

However, not all the home networking over existing wiring players are happy about the proposed G.hn standard.  MoCA and HomePlug in particular are unhappy due to incompatibility of the G.hn specification with their own proprietary technology.

One Final Word

Over the last few years, there has been an increase in the pace of development for alternate home networking technologies driven by the blurring and convergence of traditionally separate mediums.

Just a few years ago, home networking using existing wiring was more or less stalled with solutions barely capable of sharing an Internet connection, much less streaming entertainment throughout a home.  Pushed by service providers looking to provide HD quality IPTV throughout customer’s homes, many of these technologies have moved forward and matured into fast and reliable standards.  For those who are familiar with any older versions of the technology and remember the limitations should consider looking again.  They all have the potential to be good options for anyone looking to provide reliable connectivity through a home with a minimum of trouble.

References

Park, Network performance, An Overview of Key Concepts, Computer Science 422, Purdue University

Bicket, J.C., Bitrate Selection in Wireless Networks, Masters Thesis, Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology

International Telecommunication Union: Recommendation G.9954 (01/07): Home networking transceivers - Enhanced physical, media access, and link layer specifications

Multimedia over Coax Alliance: Cable and Satellite Digital Entertainment Networks

HomePlug Powerline Alliance: How HomePlug Technologies Enhance the Consumer Experience

HomePlug Powerline Alliance: HomePlug AV White Paper

HomePlug Powerline Alliance: HomePlug 1.0 Technology White Paper

Universal Powerline Association: Digital Home Specification White Paper

 

Confused about what AV Gear to buy or how to set it up? Join our Exclusive Audioholics E-Book Membership Program!

Recent Forum Posts:

DavidW posts on August 20, 2009 18:30
morrone, post: 612345
Well, I think you are missing something more basic. When the signal strength is low due to background noise, obstacles, distance, etc. the the receiver will tell the transmitter to transmit at a slower speed. The major impediment to speed is usually signal attenuation, not protocol overhead or error correction/retransmission.

I think this is nit-picking semantics.

The statement I made does not assign any proportion to various components of a signal, it simply attempts to list them. The error correction coding is a fixed ratio of any transmission or retransmission. Signal attenuation, in and of itself, leads to higher error in transmission and when significant errors are detected, retransmission is required. This is how signal attenuation affects user perceived transmission speeds.

Transceivers that are designed to drop to a lower transmission rate do so because lower bandwidth signals are typically more robust, meaning a higher percentage of the transmitted data gets through without significant errors that require retransmission. At poor enough SNR, the performance of the lower bandwidth signal can exceed that of the higher bandwidth signal which is doing more retransmitting.

Good design practice suggests that the signaling method that yields the highest actual throughput should be selected, even if it has lower theoretical throughput. In the reference section, I included a link to a paper that examines selecting various modulations to maximize throughput for variable SNR.

morrone, post: 612345
I think you misunderstand the overhead issue for ethernet too. Signal attenuation is basically a non-issue for 10/100/1000 base-T is you stay within state cable length limits. Unless there is something seriously wrong with your ethernet cable, error correction and ethernet protocol are pretty insignificant overheads. You should be able to get 95% of stated bandwidth pretty easily.

If you are only getting 70% through 10/100/1000, then it is a combination of MTU size, IP overhead, TCP overhead, OS network stack performance, and application performance. It is really not too difficult to get 90% of ethernet line rates at the application level with proper software design.

We can get about 90% of 10-GigE rates. At those higher speeds, having a fast enough bus on your motherboard and fast enough cpu and well designed application network code are far more significant factors. Ethernet overhead is basically a non-issue.

I think you are significantly overstating the ethernet overhead issue.

I never said that signal attenuation was a significant culprit in reduced Ethernet throughput, and the 70% value is likely conservative and not necessarily the value I would expect to get under all conditions; it was intended as a floor for performance.

The number is based on several sources including a Dell white paper that quoted results from a 2003 Los Alamos test of a 10GbE connection. It is fairly likely that performance has improved over the intervening years, but my intent was to provide a conservative number for goodput that had published backing.

The 70% number also likely includes issues such as data collision on congested networks under high utilization and bottle necking.

With prices for 10 GbE equipment at Newegg in the four figure range, I imagine few consumers have 10 GbE equipment yet and likely do not have a well designed professional network that avoids issues that do slow throughput over Ethernet.

The intent is to make sure no one over expects performance on a wide range of possible configurations.
ivseenbetter posts on August 20, 2009 13:40
This is a good article. It definitely raises some good points and brings these potential solutions to light for folks who may be interested. I had looked into this a few years back and the limitations were such that it didn't make sense to utilize it. However, with the info that is provided here it sounds like things are changed. I'll definitely take another look at these solutions.
morrone posts on August 19, 2009 19:13
Corrections

“For wireless transmission, the difference in maximum theoretical rate and actual rate are due to protocol overhead and error correction/retransmission. Anything that causes signal attenuation, distance and physical obstructions, will increase overhead for error correction, slowing useable data rates as a function of the signal to noise ratio with higher bit rate transmissions being more susceptible to noise.”

Well, I think you are missing something more basic. When the signal strength is low due to background noise, obstacles, distance, etc. the the receiver will tell the transmitter to transmit at a slower speed. The major impediment to speed is usually signal attenuation, not protocol overhead or error correction/retransmission.

I think you misunderstand the overhead issue for ethernet too. Signal attenuation is basically a non-issue for 10/100/1000 base-T is you stay within state cable length limits. Unless there is something seriously wrong with your ethernet cable, error correction and ethernet protocol are pretty insignificant overheads. You should be able to get 95% of stated bandwidth pretty easily.

If you are only getting 70% through 10/100/1000, then it is a combination of MTU size, IP overhead, TCP overhead, OS network stack performance, and application performance. It is really not too difficult to get 90% of ethernet line rates at the application level with proper software design.

We can get about 90% of 10-GigE rates. At those higher speeds, having a fast enough bus on your motherboard and fast enough cpu and well designed application network code are far more significant factors. Ethernet overhead is basically a non-issue.

I think you are significantly overstating the ethernet overhead issue.
davidtwotrees posts on August 17, 2009 14:11
Excellent, informative article! I'm not a geek and tend to muddle through all things techincal, and I picked up on most of the article's points. I tried wifi in my concrete shell apartment and it was terrible. I currently use a router with wired connections from my pc to my streaming blu ray player (samsung 2550), and another to my media server (escient fireball se80).
Post Reply